C. Degueldre, S. Frick
6.1 Overview
The aim of the colloid sub-programme is to understand the role of colloids in the migration of radionuclides in the geosphere. The colloid properties studied are: concentration, size distribution and nature, all of which influence their behavior under safety relevant conditions. The main activities over the past year were carried out within the framework of the Grimsel colloid project: “Colloid Formation and Migration”
(CFM), and focused on colloid generation using single particle counting (SPC) as a characterisation technique. The colloid generation approach followed is novel in this area and has the possibility of evaluating clay colloid size distributions, not only in batch systems, but also under the quasi-stagnant conditions in the CFM system. The knowledge and understanding in groundwater colloid science gained over the last decade allows well founded estimates to be made concerning the potential role played by colloids in the transport of radionuclides in the argillaceous rock formations proposed by Nagra in the frame of the Sectoral Plan for Deep Geological Disposal (SGT).
6.2 Activities in the colloid formation and migration project
The CFM project is conducted at the Grimsel Test Site (GTS), Switzerland in the framework of Phase VI of the research programme which runs from 2004 to 2013 and is dedicated to repository-relevant (i.e. large-scale, long-term) in-situ experiments. The research programme has been extended until at least 2016.
In July 2013, a radionuclide-colloid cocktail was injected into the MI shear zone at the GTS. The colloids were FEBEX bentonite doped with selected radionuclides (see Table 6.1). This tracer test (test 13-05) involved injection into the borehole CRR 99.002-i2 CFM -i2 and extraction downstream 2.23 m away at BOMI 87.010 CFM -i2 as depicted in Fig. 6.1.
The tracer cocktail was made up of three different types of tracer, AGA or G acid (7 amino-1,3-naphtalene disulfonic acid, injected mass 3706 µg, concentration 1140 µg L-1), radionuclide tracers, and bentonite colloids (injection mass 228.150 ± 5600 µg, injection concentration 70.200
± 1700 µg L-1). Table 6.1 gives an overview of the composition of the radionuclide tracer cocktail and the initial activity for each radionuclide.
The tracer cocktail was injected at a flow rate of approximately 0.3 mL min-1 (in BoMi 99.002) and water was continuously extracted at a flow rate of 5 mL min-1 from BoMi 87.010, while a flow rate of 25 mL min-1 was maintained at the instrumented surface spring on the tunnel wall (Fig. 6.1). The test was initiated by introducing the tracer cocktail into a flow loop circulating through the injection interval at a relatively high rate to keep the interval well mixed while maintaining a near constant net injection flow rate into the shear zone. The volume of the vessel containing the tracer cocktail was 2.25 L, and the volume of the injection flow loop was 1.0 L, so that the entire injection circuit volume was 3.25 L. This arrangement resulted in an exponentially decaying source term in the shear zone as the tracers were slowly blend out of the injection circuit.
Table 6.1: Composition of radionuclide tracer cocktail (in 3.25 L) in CFM Tracer Test 13-05.
Nuclide Activity at t(0)*
(Bq)
Colloidal Fraction Nuclide Activity at t(0)*
(Bq)
Colloidal Fraction
22Na 1.50 ∙ 10+06 0-0.03 237Np 1.21 ∙ 10+02 1.0
133Ba 9.60 ∙ 10+06 0.24-0.34 243Am 1.74 ∙ 10+03 1.0
137Cs 9.00 ∙ 10+05 0.97 242Pu 1.67 ∙ 10+02 1.0
*t(0): time 0, the time of cocktail preparation
The identity and activities of the tracers in the injection cocktail, as well as the fractions of each radionuclide that were initially partitioned to the bentonite colloids in the cocktail, are listed in Table 6.1. The tracer concentrations in the extracted water as a function of time are shown in Fig. 6.2. All the data shown in Fig. 6.2 should be considered preliminary. The actinide concen-trations were also measured by ICP-MS at the Karlsruher Institute of Technology, KIT. The 22Na,
137Cs and 133Ba activities were measured by gamma spectrometry at both PSI and KIT, and were found to be in excellent agreement. The Amino-G acid concentrations were measured in the field using an inline fluorimeter and were found to be in good agreement with offline measurements.
Fig. 6.1: Schematic illustration of the CFM field test zone. Injection in BoMi 99.002 and breakthrough downstream at BoMi 87.010, within the flow oriented toward the surface spring in the access tunnel wall.
Fig. 6.2: Comparison of 22Na, 137Cs and 133Ba (Bq) and AGA (µg L-1) breakthrough curves in the CFM tracer test 13-02 at the Grimsel Test Site.
Offsite colloid measurements were conducted by PSI using a single-particle counter (SPC) and the data are shown in Fig. 6.3. The colloid concen-trations were also measured in the field using a mobile laser-induced breakdown detection (LIBD) system operated by KIT personnel. The general
trends in the LIBD and SPC data were in good agreement. The SPC data are size normalized.
They show that the colloid size ranges relevant for the breakthrough are between 50-100 and 100-150 nm, and that sizes above 200 nm have little effect on the transport in the test.
The activity and concentration maxima were observed between 2160 and 2520 mins for the colloids (independent of size), for AGA, and for Cs and Ba; and between 2880 and 2940 mins for Na.
The maximum for Na appears a little later because of its weaker sorption on the colloids (see Table 6.1) and also because it sorbs weakly on the host rock within which it may diffuse deeper than the colloids. The analysis of the data is in progress in close cooperation with the CFM modelling group.
Fig. 6.3: Comparison of colloid breakthrough curves in the CFM tracer test 13-02 at the Grimsel Test Site. Conditions: single particle counting (SPC) data recorded for the 50-100, 100-150, 150-200, 200-300, 300-500, 500-700 and 700-1000 nm size range channels.
6.3 Other colloid activities
The co-operation with CIEMAT continues in a very productive manner. The main aim is to optimise the radio-analytical effort required to measure colloid breakthrough curves in the CFM experiment at the GTS.
The application of the advances made in groundwater colloid sciences over the last decade have allowed colloid data to be derived for the hydro-geochemical systems in the argillaceous rock formations proposed by Nagra in the frame of the SGT. A publication has been prepared which includes:
• groundwater colloid concentration results from field experiments ranging from dilute systems to saline groundwaters,
1.E-01 1.E+00 1.E+01 1.E+02
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05
Time (min)
Activity (Bq) & Conc.(µg L-1) Na-22
Cs-137
1.E+01 1.E+02 1.E+03 1.E+04 1.E+05
Time (min)
Normalized Concentration (ml-1 nm-1) 50
100
• a comparison of colloid concentration results with laboratory batch experiments, and
• the results of colloid adhesion tests (attachment factor values).
The study will be completed by calculations of colloid concentrations in the relevant systems using the suspension pseudo-equilibrium model (DEGUELDRE et al., 2009). This field/lab and model study will provide the required colloid data for the rock formations being investigated by Nagra.
The new model of colloid generation (batch stagnant & transient) has also been applied to upgrade the Bangombe (fossil reactor in Gabon) colloid data produced in 1996.
6.4 Future work
At Grimsel, the clay source foreseen for the long term CFM experiment will be installed early in 2014, and intensive sample analysis campaigns are anticipated in 2014-2015. A PhD student from CIEMAT will be involved in the analytical effort required (mainly) at the beginning of the new CFM experiment. The colloid project leader has been asked to participate at the CFM Modellers & Lab meeting early in February 2014 at the NDA, Oxford and at the CFM meeting in Asia the autumn of 2014. Participation at the BelVar meeting in Meringen (16-18 June 2014) is also anticipated.
6.5 References
DEGUELDRE C., AEBERHARD PH., KUNZE P., BESSHO K.(2009)
Colloid generation/elimination dynamic processes:
Toward a pseudo-equilibrium? Colloids Surf. A, 337, 117-126.